Hofstra Law ReviewVolume 40Issue 1 FORTIETH ANNIVERSARY VOLUME Article 10

2011

The Pari Passu Clause in Sovereign BondContracts: Evolution or Intelligent Design?Mark L.J. Wright

Follow this and additional works at: http://scholarlycommons.law.hofstra.edu/hlr

Part of the Law Commons

This document is brought to you for free and open access by Scholarly Commons at Hofstra Law. It has been accepted for inclusion in Hofstra LawReview by an authorized administrator of Scholarly Commons at Hofstra Law. For more information, please contact lawcls@hofstra.edu.

Recommended CitationWright, Mark L.J. (2011) "The Pari Passu Clause in Sovereign Bond Contracts: Evolution or Intelligent Design?," Hofstra Law Review:Vol. 40: Iss. 1, Article 10.Available at: http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

THE PARI PASS U CLAUSEIN SOVEREIGN BOND CONTRACTS:

EVOLUTION OR INTELLIGENT DESIGN?

Mark L. J. Wright*

I. THE PARI PASSU CLAUSE: Coccyx, THIRD MOLARS, OR THYMUS?

In 1893, Robert Wiedersheim published a book on human anatomythat claimed to identify eighty-six human organs that had been renderedfunctionless by the process of evolution.' A century of subsequentresearch has confirmed Wiedersheim's findings for some organs and hasoverturned others. It is now widely agreed that some parts of the body,such as the coccyx (the tailbone), are functionless but relativelyharmless.2 Others, like third molars (wisdom teeth), no longer serve auseful purpose and in some cases have become harmful and need to besurgically removed.3 Still others have turned out to play important rolesin the body, such as the thymus gland, which helps to regulate theimmune system.4

Much of the recent debate surrounding the pari passu5 clause insovereign bonds has concluded that it is at best a coccyx and essentiallymeaningless, and at worst, a set of wisdom teeth that introducesunnecessary litigation risk and needs to be surgically removed from

* Senior Economist, Federal Reserve Bank of Chicago, Assistant Professor, University of

California, Los Angeles, and Faculty Research Fellow, National Bureau of Economic Research. Theauthor would like to thank Mitu Gulati for many illuminating discussions, and the editors for theircomments. The views expressed herein are those of the author and not necessarily those of theFederal Reserve Bank of Chicago or the Federal Reserve System.

1. See generally R. WIEDERSHEIM, THE STRUCTURE OF MAN: AN INDEX TO His PAST

HISTORY (G. B. Howes ed., H. and M. Bernard trans., London, MacMillan & Co. 2d ed. 1895)(classifying certain organs as vestigial in nature).

2. See J. SPEED ROGERS ET AL., MAN AND THE BIOLOGICAL WORLD 313 (1942).

3. Id.4. Walter R. Benson, The Thymus, 199 JAMA 1013, 1013 (1967) (book review).5. See MITu GULATI & ROBERT E. SCOTT, THE THREE AND A HALF MINUTE TRANSACTION:

BOILERPLATE AND THE LIMITS OF CONTRACT DESIGN (forthcoming 2012) (manuscript at 21) (onfile with the Hofstra Law Review).

1

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA LAW REVIEW

future sovereign bond contracts (such is the position taken in themonograph by Mitu Gulati and Robert E. Scott,6 which served as thestarting point for this symposium). By contrast, in this Article I willargue that the clause is more aptly likened to the thymus gland-a clausewhose meaning is not well understood and whose origin is perhaps lostin history, but that nonetheless fulfills a potentially important role that islikely to expand further in the future.

I begin by reviewing the historical context in which the introductionof the pari passu clause into sovereign bonds took place. I argue that thisperiod was one in which concerns for fair and equitable treatment ofcreditors in the event of a sovereign debt restructuring were central todiscussions between creditors and sovereigns, and that the introductionof the clause was a rational response to this environment; that is, the paripassu clause was the result of more or less intelligent design by debtorsand creditors. I then review the evolution of concerns for the fair andequitable treatment of creditors over time, the evolving interpretation ofwhat constitutes fair treatment, and the ways in which the term paripassu has been used by creditors.

I also argue that the growing use of the pari passu clause today, andthe drift towards the "risky" version of the clause specifying pro ratarepayment, is the rational response to expectations of future contractualand institutional changes that are likely to intensify concerns for inter-creditor equity and fairness in the future. I conclude with somereservations about Gulati and Scott's use of qualitative survey evidenceto infer the rationality of actions.

II. WAS THE SOVEREIGN PARI PASSU CLAUSE THE RESULT OF

INTELLIGENT DESIGN?

As documented by Gulati and Scott, the first sovereign bondcontract to contain a pari passu clause was issued by Bolivia in 187 1.7

This places the genesis of the contract at a point in time that is importantfor at least two reasons. First, it establishes that the contract wasintroduced at the beginning of what is known as the "Golden Age" ofinternational capital markets.8 One of the key distinguishing features ofthis period is that sovereign lending took place in the shadow of theabsolute version of the Doctrine of Sovereign Immunity (the

6. See generally id.7. Id. (manuscript at 152).8. Paolo Mauro et al., Emerging Market Spreads: Then Versus Now, 117 Q.J. ECON. 695,

695-96 (2002) (stating that the previous golden age for emerging market bonds and internationalcapital flows was from 1870-1913).

[Vol. 40:103

2

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

EVOLUTION OR INTELLIGENT DESIGN?

"Doctrine"). 9 Among other things, the Doctrine made it impossible for asovereign to credibly pledge its assets as collateral against a loan,notwithstanding the existence of such pledges and earmarks in sovereignbonds.'0 Combined with the fungible nature of sovereign revenuestreams, this meant that all sovereign loans were de facto claims againstthe future general revenues of the sovereign backed only by its "full faithand credit."' 1 In such an environment, the only purpose of a contract thatis not enforceable is to make public the expectations of the parties to theagreement so that it defines, in the event of future repayment difficulties,the boundaries of negotiations to restructure debts. It is in this contextthat the contractual terms of sovereign lending must be interpreted.

In the absence of legal enforcement, the primary incentive for acountry to repay its debts was to preserve future access to creditmarkets.' 2 The second reason why the 1871 origin of the clause isimportant is that it places its introduction at a pivotal time in theevolution of the mechanisms by which competing creditors interacted innegotiating with sovereign countries in default, and acted to limit futurecredit market access. For much of the nineteenth century, the largest andmost liquid market for sovereign debt was in London. 13 In 1826, theLondon Stock Exchange adopted a rule that prevented the listing of newbond issues by sovereign countries that were in default and had notreached a settlement with its creditors:

The committee will not sanction or recognize bargains made in newbonds, stock, or other securities issued by any foreign Government thathas not duly paid the dividends on former loans raised in this country,unless such Government shall have effected and carried out asatisfactory arrangement with the holders of such stock, bonds, or othersecurities, on which the dividends have been left in arrear.14

9. See Ugo Panizza et al., The Economics and Law of Sovereign Debt and Default, 47 J.ECON. LrrERATURE 651, 653 (2009).

10. See I EDWIN BORCHARD, STATE INSOLVENCY AND FOREIGN BONDHOLDERS 81-82

(1951).11. Id. at 82 (internal quotation marks omitted).12. GULATI & SCOTr, supra note 5 (manuscript at 30-3 1).13. PAOLO MAURO ET AL., EMERGING MARKETS AND FINANCIAL GLOBALIZATION:

SOVEREIGN BOND SPREADS IN 1870-1913 AND TODAY 134 (2006); Mauro et al., supra note 8, at

698-99.14. Money-Market & City Intelligence, TIMES (London), Nov. 23, 1868, at 8 (internal

quotation marks omitted). See also I BORCHARD, supra note 10, at 172-73; Larry Neal & LanceDavis, Why the London Stock Exchange Dominated the First Age of Globalization, 1801-1914, at19 (Oct. 1, 2005) (unpublished manuscript), available at http://www.cepr.org/meetstwkcn/l/l617/papers/neal.pdf.

2011]

3

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA LAW REVIEW

The consequence was to make it if not impossible then at least moreexpensive for a sovereign country in default to obtain new loans in theLondon market until it had reached a satisfactory settlement innegotiation with its creditors.

In practice, the London Stock Exchange looked to creditorsthemselves for guidance as to whether or not a country in default hadmade a "satisfactory arrangement."'15 In the first few decades followingthe adoption of this rule, creditors in London would typically organizethemselves into bondholder committees for the purposes of thesenegotiations. 16 The absence of a formal system, or any overarchingcreditor body, meant that it was not uncommon for rival creditorcommittees to form with competing settlement proposals. 17 Followingthe completion of negotiations with one creditor committee, individualbondholders were asked to register their acceptance of the agreementwith either the committee or the sovereign.' 8 This complicated thenegotiation process for the sovereign (which had to bargain withmultiple creditor groups), the London Stock Exchange (which had todetermine which, if any, of several proposed settlements wassatisfactory), and bondholders (who were reluctant to accept one set ofterms when a rival committee might eventually secure better ones).

By the 1860s, motivated by the problem of rival committees, effortsto form a unified creditor organization were well underway andculminated in the 1868 formation of the Council of ForeignBondholders, which became the Corporation of Foreign Bondholders("CFB") in 1873.19 Although initially envisaged as a body that wouldrepresent bondholders operating in all of the main creditor nations, theCFB was finally constituted to restrict itself to the representation ofBritish bondholders, although also undertaking to coordinate withsimilar bodies in the other continental European markets.2°

By the time of the Bolivian bond issue that included the first paripassu clause, concerns about conflict among bondholder groups, and thepossibility of discriminatory settlements, were still very much in effect.21

15. BARRY EICHENGREEN ET AL., CRISIS? WHAT CRISIS? ORDERLY WORKOUTS FOR

SOVEREIGN DEBTORS 20 (1995).16. Id. at 19.17. Id. at 19-20.18. Id. at 20.19. 1 BORCHARD, supra note 10, at 203, 205.20. See MAURO ET AL., supra note 13, at 150.21. See Rui Pedro Esteves, Quis Custodiet Quem? Sovereign Debt and Bondholders'

Protection Before 1914, at 6-7 (Univ. of Oxford Dep't of Econ. Discussion Paper Series, Paper No.323, 2007), available at http://www.economics.ox.ac.uk/Research/wp/pdf/paper323.pdf. But seeMAURO ET AL., supra note 13, at 148.

[Vol. 40:103

4

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

EVOLUTION OR INTELLIGENT DESIGN?

In 1871, the long term viability of the Council of Foreign Bondholderswas far from assured, and there was no mechanism for ensuring fair and

- 22equitable settlement terms for bondholders in different countries. As aresult, it seems reasonable to conjecture that the introduction of the paripassu clause was intended, at least in part, as a tool for preserving inter-creditor equity and fairness in negotiations.

III. THE EVOLVING CONCEPT OF FAIRNESSAND THE PARI PASSU CLAUSE

After 1868, the London Stock Exchange deferred to the CFB in23determining whether a settlement with a country was satisfactory.

There is a great deal of material written about various specificnegotiations, and about the nature of the debate on these settlements.However, it is not always obvious how much weight should be attachedto various public arguments since different creditors appeared to beprepared to raise any argument necessary to ensure the most favorablesettlement terms. Nonetheless, by looking at these materials we candiscern some patterns as to whether certain settlement features wereviewed by creditors as satisfactory.

One thing that is clear is that a satisfactory or fair settlement did notrequire that all creditors be treated proportionately. Recognizing thesovereign's right to tax income within its borders, creditors oftenconceded the right of the sovereign to treat domestic debts as junior toforeign debts (for example, Austria in the 1860s and the DominicanRepublic in the 1900s).24 Likewise, "funding debts"--debts that hadbeen issued in a past debt restructuring-were often given priority on thegrounds that creditors should not have their already reduced claimsfurther reduced in value (for example, Greece in the 1930s).25 Althoughthere are some exceptions, it was also common for debts issued withsome form of security or pledge to receive more favorable terms in

26negotiations, 6 while debts issued at different times received priority insome cases (for example, Egypt in the 1870s and Turkey in the 1880s).27

22. See MAURO ET AL., supra note 13, at 132.

23. EICHENGREEN ET AL., supra note 15, at 20.24. See I BORCHARD, supra note 10, at 359-60, but see id. for a list of exceptions to this

principle. See also Money-Market & City Intelligence, TIMES (London), Mar. 10, 1868, at 10("Those who buy the internal stock of Austria must be prepared to conform to whatever fate awaitsthe Austrian public, but a coupon payable abroad in sterling should always be good for the amount itexpresses on its face.").

25. 1 BORCHARD, supra note 10, at 338-42, 357 n.62.26. Id. at 356-57.27. Id. at 343, 344 & n.22.

2011]

5

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA LAW REVIEW

In most of these cases, priority took the form of a greater claim on thegeneral funds available for repayment, independently of whether or notany specific revenue source or asset had been pledged.28

There was also widespread agreement that certain forms ofdiscrimination were not fair, the most notable of which beingdiscrimination on the basis of the nationality of the creditor.29 However,even where equal treatment was demanded, there was not alwaysconsensus as to what that meant. For example, during Chile'srestructuring in the 1930s, British creditors protested at plans to pay acommon but reduced rate of interest on all bonds (that were issued withdifferent interest rates), insisting on a proportionate reduction in theinterest rates on all bonds.30 This proportionate rule seems to havebecome common thereafter.

Over time, concerns for inter-creditor equitable treatment have notdiminished, and there is some evidence that they have convergedsomewhat on an insistence for pro rata repayment.31 Although theInternational Monetary Fund demands priority of repayment on its loans,official sector lenders such as creditor country governments restructuringdebts under the Paris Club routinely insist on comparable (or at least, noless favorable) treatment to private creditors, which is typicallyinterpreted on a proportional basis.32

Private creditors have increasingly written clauses into sovereigndebt contracts designed to constrain the ability of a sovereign to makediscriminatory repayments. These include negative pledge clauses toensure that the debtor will not pledge any of its assets to subsequentcreditors, mandatory prepayment clauses requiring pro rata payments toall lenders in the event of a prepayment to any lender, and cross-defaultclauses allowing any lender to declare a loan in default should the debtordefault on any other loan and so prevent early defaulting loans fromreceiving better terms.33 The typical syndicated sovereign loan of the1970s and 1980s went so far as to include a sharing clause that explicitlyensured pro-rated payments in the event of a default. 34

28. See id. at 356.29. Id. at 338.30. COUNCIL OF THE CORP. OF FOREIGN BONDHOLDERS, SIXTY-SECOND ANNUAL REPORT 27-

32 (1935) [hereinafter SIXTY-SECOND ANNUAL REPORT].31. See GULATI & SCOTT, supra note 5 (manuscript at 23).32. NOURIEL ROUBINI & BRAD SETSER, BAILOUTS OR BAIL-INS? RESPONDING TO FINANCIAL

CRISES IN EMERGING ECONOMIES 250-51,256-57 (2004).33. GULATI & SCOTT, supra note 5 (manuscript at 21 & n.16, 34, 36, 89).34. Lee C. Buchheit, Changing Bond Documentation: The Sharing Clause, INT'L FIN. L.

REV., July 1998, at 17, 17.

[Vol. 40:103

6

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

EVOLUTION OR INTELLIGENT DESIGN?

Set against this history, it seems likely that the pari passu clausewas intended to be a component of a system designed to ensure inter-creditor equity, possibly as a substitute for these other clauses (inparticular, the negative pledge clause), or perhaps even as a complementto them. To refine our understanding of the intention of creditors inintroducing and using these clauses, we can look to various forms ofevidence-both direct and indirect-as to creditors' interpretations ofthe pari passu clause.

Indirect evidence that the pari passu clause is viewed as a substitutefor the negative pledge clause comes from Gulati and Scott's data oncontractual terms. According to the data, of the 275 pre-World War IIbonds issued that could be interpreted as being unsecured, only threecontained both negative pledge and pari passu clauses." Moreover,during the modem period in which the interpretation of the negativepledge clause has been broadened to address preferential arrangementsthat do not create formal security interests, the pari passu clause hasbeen weakened.36

Indirect evidence that the pari passu clause is viewed as acomplement to these other clauses, and in particular as a clause designedto ensure pro rata payments, comes from its use following the BrusselsCourt of Appeal decision in Elliott Associates, L.P. 37 As shown byGulati and Scott, since Elliott, the clause has become more prevalent ininternational bonds, and moreover the version of the clause that favors a

38pro rata payment interpretation has become more widespread. Directevidence in favor of the pro rata interpretation comes from the fact thatthis interpretation has been used explicitly by creditors in responding toEcuador's 2009 selective default.39

Ideally, we would like to have direct qualitative evidence ofcreditor attitudes to the clause along the lines of the qualitative evidencepresented by Gulati and Scott on the attitudes of creditors' legalrepresentatives. Gulati and Scott report some evidence that lawyers

35. GULATI & ScoTr, supra note 5 (manuscript 141-42). Albeit only ten of the pre-WorldWar II bonds classified restrictively as "unsecured" possessed a pari passu clause, Gulati and Scottdo not report the fraction of these bonds that possessed a negative pledge clause. See id. (manuscriptat 146, 199 tbl.A7).

36. See id. (manuscript at 143-44).37. General Docket No. 2000/QR/92 (Court of Appeal of Brussels, 8th Chamber, Sept. 26,

2000) (unofficial translation on file with the Hofstra Law Review).38. GULATI & ScOrr, supra note 5 (manuscript at 87, 185-86).39. Id. (manuscript at 185).

2011]

7

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA L4W REVIEW

perceive their clients' interpretation to be one supporting pro ratapayment:

Can you imagine the negotiation dynamic? It doesn't matter that thecreditors themselves disagree with Elliott. They like the notion of inter-creditor parity... you know, equal treatment. How do I go to thecreditors and say that I want to remove a clause that says that all of youwill be treated equally?40

Further qualitative evidence comes from examining creditordiscussions of the clause in history. A quick review of the AnnualReports of the CFB reveals four instances in which the term pari passuwas used. In three of these cases, the clause is used to denote a de jure orde facto claim for proportional repayment. In the case of the Austriangovernment's repayment of the German 1929 First Mortgage SterlingBonds, the term is used explicitly to specify a proportional distributionof payments between two bonds (although other bonds were givenpriority).41 In the case of the 1935 Province of Buenos Aires debtrestructuring, the term is used to signify an equal claim on the fundsavailable for the repayment of the debt.42 In the Greek Public WorksSterling Loan of 1931, the government reserved the right to issue futuredebt that would rank pari passu for security, where that security took theform of the "surplus revenues" of the government.43 Only in the case ofthe 1933 Colombian debt restructuring is the term used to denote anequal ranking of claims (for British and U.S. bondholders) withoutspecific mention of payment.44

IV. WHY WAS THE PARI PASSU CLAUSE NOT SURGICALLY REMOVEDFOLLOWING ELLIOTT?

I have argued above that the introduction of the pari passu clausewas the result of more-or-less intelligent design by a debtor and itscreditors in an environment in which concerns as to the fair treatment ofdifferent creditors in a debt restructuring were particularly active.Discussions of the clause by creditors following its introduction showthat it has also evolved, taking on differing meanings at different times-at one point appearing to substitute for negative pledge clauses, and at

40. Id. (manuscript at 173).41. See COUNCIL OF THE CORP. OF FOREIGN BONDHOLDERS, EIGHTY-SECOND ANNUAL

REPORT 22-23 (1955).42. See SIXTY-SECOND ANNUAL REPORT, supra note 30, at 20-21, 96.43. COUNCIL OF THE CORP. OF FOREIGN BONDHOLDERS, SIXTIETH ANNUAL REPORT 240-41

(1933).44. See id. at 30-31, 161-62.

[Vol. 40:103110

8

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

EVOLUTION OR INTELLIGENT DESIGN?

other points complementing them by acting as a guarantee of at leastequally favorable treatment, and in some cases explicitly proportionaterepayment.

In this light, the problem that creditors and regulators have with theElliott decision seems to be not so much with the pro rata interpretationof the clause, but with its application in that specific case. The creditorsthat had participated in the Brady bond45 exchange with the Republic ofPeru presumably would have been quite content if the pari passu clausewere used to ensure that Elliott Associates, L.P. received the sameproportional reduction in payment that they themselves had received inthe exchange.46 However, in combination with the fact that Peru wasservicing the new but already-reduced claims under the Brady bonds infull, the creditors objected to the use of the clause in order to obtain fullrepayment of the original unreduced bonds.47

Nonetheless, given the risk of litigation and the potential continuedproblematic application of the principle, why was the clause not deleted,or at least modified to reduce this problem, in succeeding bond issues? Isuspect that there were three reasons for the continued usage of theclause and the move towards the more explicit pro rata version of theclause. First, creditors continued to desire the protection againstdiscriminatory settlements that the pari passu clause was perceived asproviding. Indeed, the importance of this protection was illustrated withthe discriminatory default by Ecuador in 2009.48 Second, creditorsbelieved (correctly) that changes in the operation of Euroclear and otherpayment-clearing mechanisms would prevent the clause from being usedto block the repayment of funding bonds in the future.4 9

Third, and perhaps most importantly, creditors anticipated that therewould be further changes in sovereign bond contracts and/orinternational institutions that would work to limit the misapplication ofthe pro rata interpretation, and that would make pro rata protectionmore important in the future. Creditors were correct in the former belief,

45. Brady bonds were developed by former U.S. Treasury Secretary Nicholas Brady torestructure the debt of developing countries. FED. RESERVE BD. OF GOVERNORS, TRADING &

CAPITAL-MARKETS ACTIVITIES § 4255.1 (1998), available at http://www.federalreserve.gov/Boarddocs/SupManualltrading/4000p2.pdf.

46. But see Transcript of Oral Argument at 12, 14, NML Capital, Ltd. v. Republic ofArgentina, 2011 WL 4529332 (S.D.N.Y. Feb. 23, 2012) (No. 08 Civ. 6978 (TPG)) where thisargument was raised but dismissed on the grounds that the Argentine restructuring agreement was a"voluntary exchange[] with sophisticated purchasers of bonds" which "gave participants in thatexchange offer the right to participate in any future exchange offers but not the right to receive thesame treatment as a bondholder that settles a lawsuit with Argentina."

47. See GULATI & ScOTr, supra note 5 (manuscript at 21).48. See id. (manuscript at 183-84).49. See id. (manuscript at 57, 192).

2011I]

9

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA LAW REVIEW

with the widespread adoption of collective action clauses in sovereignbond contracts making it possible, in principle, to impose commonsettlement terms on all holders of a given bond through a super-majorityvote of bondholders, thus reducing the litigation risk from holdoutcreditors.50

Creditors may yet turn out to be correct in the latter belief as well.Following the example of Uruguay,5' there has been some discussion ofthe possibility of the more widespread use of aggregation clauses thatwould allow the imposition of settlement terms on the holders of otherbonds, provided that a proposed settlement be accepted by a super-majority of the holders of each bond, as well as a super-majority of allbondholders voting collectively. This discussion has intensified in thecontext of the current European debt crisis following the commitment bythe Eurogroup to introduce aggregation clauses into all European Union

52area bonds starting in 2013, which might lead to their more widespreadadoption by other countries. The adoption of aggregation clauses opensthe possibility that future debt restructuring operations will involve amore heterogeneous group of creditor claims than previously observed.In such an environment, the risk of a discriminatory settlement isenhanced, and the protection offered by a pro rata payment clause islikely to be more valuable.

V. WHAT SHOULD WE MAKE OF THE SURVEY EVIDENCE?

I conclude with some remarks on the use of qualitative surveyevidence to infer the rationality, or irrationality, of actions. Gulati andScott begin their piece with the "story of the jumping frog": a parabledesigned to illustrate the dangers associated with inferring themotivations for, and rationality of, a set of actions from data on theactions themselves alone.53 Gulati and Scott aim to surmount thisproblem by surveying market participants directly as to the reasoningunderlying their actions. 4 Confronted by an inability or unwillingness toarticulate these reasons, the actions are interpreted as being irrational,and their persistence is attributed to the workings of the organizationswithin which the actions take place.55

50. See id. (manuscript at 37-38).51. See id. (manuscript at 169).52. Press Release, Eurogroup, Eurogroup Statement on European Financial Stabilization 1-2

(Nov. 28, 2010), http://www.consilium.eumpa.eu/uedocs/cms-data/docs/pressdata/en/ecofin/l18050.pdf.

53. See GULATI & SCOTT, supra note 5 (manuscript at 7).54. See id. (manuscript at 10- 11).55. See id. (manuscript at 12, 121).

[Vol. 40:103

10

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10

EVOLUTION OR INTELLIGENT DESIGN?

While surveys that attempt to collect data on an agent's motivationhave the potential to be very valuable, there remains a degree of hostilityto their use within some branches of the social sciences. This is for atleast two reasons. First, it is not always obvious that respondents have anincentive to answer truthfully. In the present context, there are at leasttwo reasons, implicitly pointed out by Gulati and Scott themselves, whyrespondents might have an incentive to engage in ex post rationalizationof their actions. For one, the survey itself could be interpreted ascriticism of the lawyers' own drafting practice which may bias theiranswers. 56 For another, given the hostility of their clients to the Brusselsinterpretation, "the lawyers on the sovereign side had to conjure up ameaning for the clause (it would hardly do [anything] for them to bearguing to a judge that they knew that Elliott was wrong but that theyhad no clue as to what the clause actually meant). 57

The second reason for hostility is that, even if we accept thatrespondents have an incentive to answer truthfully, it is not entirelyobvious that agents need to be able to explain the rational basis of theiractions for their behavior to be rational. This argument is reminiscent ofan old one due to Milton Friedman and Leonard J. Savage. 58 In thatcontext, Friedman and Savage were responding to criticism of theeconomist practice of modeling human behavior as the (rational)outcome of a complicated mathematical optimization problem.59 Theyused the example of a billiard player to argue that there was no usefulempirical distinction between an agent that solved complicatedmathematical problems as part of his or her decision making process,and one that merely acted as if he or she solved the same mathematicalproblems.6 °

Consider the problem of predicting, before each shot, the direction oftravel of a billiard ball hit by an expert billiard player. It would bepossible to construct one or more mathematical formulas that wouldgive the directions of travel that would score points and, among these,would indicate the one (or more) that would leave the balls in the bestpositions. The formulas might, of course, be extremely complicated,since they would necessarily take account of the location of the balls inrelation to one another and to the cushions and of the complicatedphenomena introduced by "english." Nonetheless, it seems not at all

56. See id. (manuscript at 126-27).57. Id. (manuscript at 188).58. See Milton Friedman & L. J. Savage, The Utility Analysis of Choices Involving Risk, 56 J.

POL. ECON. 279, 297-98 (1948).59. See id. at 287-88, 297-98.60. Id. at 298.

20111

11

Wright: The Pari Passu Clause in Sovereign Bond Contracts: Evolution or I

Published by Scholarly Commons at Hofstra Law, 2014

HOFSTRA LAW REVIEW

unreasonable that excellent predictions would be yielded by thehypothesis that the billiard player made his shots as if he knew theformulas, could estimate accurately by eye the angles, etc., describingthe location of the balls, could make lightning calculations from theformulas, and could then make the ball travel in the direction indicatedby the formulas. It would in no way disprove or contradict thehypothesis, or weaken our confidence in it, if it should turn out that thebilliard player had never studied any branch of mathematics and wasutterly incapable of making the necessary calculations: unless he wascapable in some way of reaching approximately the same result as thatobtained from the formulas, he would not in fact be likely to be anexpert billiard player.61

In the present context, it should in no way disprove or contradict thehypothesis that expert sovereign debt practitioners are acting rationally if

they are utterly incapable of articulating the rational basis for their

actions; unless they are acting rationally in drafting these contracts, theywould not be likely to be expert sovereign debt practitioners.

61. Id.

[Vol. 40:103

12

Hofstra Law Review, Vol. 40, Iss. 1 [2014], Art. 10

http://scholarlycommons.law.hofstra.edu/hlr/vol40/iss1/10